> ## Documentation Index
> Fetch the complete documentation index at: https://api-tools.memories.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Create Collection

> Create a collection

<Info>
  **Product**: Video Datalake
  **Host**: `https://api.memories.ai/datalake/v1`
  **Auth**: `Authorization: sk-mai-...`
</Info>

Create a **collection** — the container that fixes the safety detectors and face-recognition switch that every video uploaded into it inherits. Indexing always uses **OmniRetriever**, our multilingual video-embedding model (3072-dim); there is no model to choose.

<Note>
  **Pricing:** Free — management / read (no charge). · See [Pricing](/datalake/pricing) for the full model.
</Note>

## Response Fields

<ResponseField name="id" type="string">Collection ID (`col_` prefix).</ResponseField>
<ResponseField name="embedding_model" type="string">The embedding model used to index this collection — always `OmniRetriever`.</ResponseField>
<ResponseField name="video_count" type="integer">Live count of videos in the collection.</ResponseField>
<ResponseField name="face_recognition_enabled" type="boolean">Whether persons/entities are enabled.</ResponseField>


## OpenAPI

````yaml POST /datalake/v1/collections
openapi: 3.1.0
info:
  title: Memories.ai DataLake API 2.0
  version: 2026-08
  description: OpenAPI mapping used by Mintlify Try it for the DataLake API 2.0 docs.
servers:
  - url: https://api.memories.ai
security:
  - ApiKeyAuth: []
paths:
  /datalake/v1/collections:
    post:
      tags:
        - Collections
      summary: Create a collection
      description: >-
        Create a **collection** — the container that fixes the safety detectors
        and face-recognition switch that every video uploaded into it inherits.
        Indexing always uses **OmniRetriever**, our multilingual video-embedding
        model (3072-dim); there is no model to choose.
      operationId: create_collection
      requestBody:
        required: true
        content:
          application/json:
            schema:
              type: object
              properties:
                name:
                  type: string
                  description: Display name, ≤255 chars
                enabled_detectors:
                  type: array
                  items:
                    type: string
                    enum:
                      - safety_detector
                  description: Detector allow-list. Currently only `safety_detector`.
                face_recognition_enabled:
                  type: boolean
                  description: Master switch for persons/entities. Default false.
                metadata:
                  type: object
                  description: >-
                    Free-form. Face-compliance docs can go under
                    `consent_document`.
              required:
                - name
              example:
                name: demo
                enabled_detectors:
                  - safety_detector
      responses:
        '200':
          description: Success
          content:
            application/json:
              example:
                id: col_xxx
                name: my-set
                embedding_model: OmniRetriever
                enabled_detectors: []
                face_recognition_enabled: false
                video_count: 0
                metadata: {}
                reid_threshold: null
                person_threshold: null
                created_at: '2026-07-18T02:00:00Z'
components:
  securitySchemes:
    ApiKeyAuth:
      type: apiKey
      in: header
      name: Authorization
      description: 'Send `Authorization: sk-mai-...`'

````